By the way, those are present epoch years of 8766 hours. A billion years ago, there were fewer hours in a year.

Shouldn't that be **more** hours? (assuming that previous epoch hours are still defined as day/24).

Rotation slowing due to tidal effects --> day lengthening --> shorter day (and hour) in previous epochs.

Orbital expansion or contraction affecting the year is surely smaller than this effect?

You're right, and I'm wrong. I was mistakenly assuming that the hour is defined as 86,400 seconds; in fact, the hour is defined as 1/24 of a mean tropical day. The length of the year in seconds is constant, and the length of the day in hours is constant, but the length of day in seconds is increasing. A billion years ago, the year had more days, and therefore more hours.

The present rate angular acceleration of Earth's rotation due to tidal friction is about .000023 s/year, or 23,000 sec/billion year. The rate of acceleration is slowing, but there's no way to know how fast it has slowed in the last billion years because tidal friction changes as ice ages come and go. We can say that a billion years ago the day was less than 63,000 seconds. Perhaps it was half of the present 86,400 seconds.

I should have given my result as 37,800 seconds, rather than 10.5 hours. That way, I wouldn't have to specify that I meant 10.5 of today's hours. At any rate, the difference is many orders of magnitude less than the margin of error in measuring the age of crystals.